Land Cover Change

Land cover change is defined as the loss of natural areas, particularly loss of forests to urban or exurban development, or the loss of agricultural areas to urban or exurban development.

From: Anthropocene , 2018

Urban ecology – current state of research and concepts

Pramit Verma , ... A.S. Raghubanshi , in Urban Ecology, 2020

3.3 Land use land cover change

Land cover change denotes a change in certain continuous characteristics of the land such as vegetation type, soil properties, and so on, whereas land-use change consists of an alteration in the way certain area of land is being used or managed by humans ( Patel et al., 2019). This involves the transformation in the natural landscape due to urban growth. It is interesting to note that this change is responsible for a number of local and global effects, including biodiversity loss and its associated effects on human health, and the loss of habitat and ecosystem services (Patel et al., 2019). It is mainly driven by urban growth and is particularly important now for developing and underdeveloped countries. However, natural causes may result in land cover change, but land-use change requires human intervention (Joshi et al., 2016).

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978012820730700001X

Modeling of coastal vulnerability to sea-level rise and shoreline erosion using modified CVI model

Kaliraj Seenipandi , ... N. Chandrasekar , in Remote Sensing of Ocean and Coastal Environments, 2021

4.9 Land use and land cover change

LULC change regulates the availability of sediments to coastal landforms due to the controlling erosion and accretion process (Rao, 2002; Bradley, 2009; Butt et al., 2015). The LULC features are extracted using Landsat ETM+   images (30   m) acquired in 2000 and 2018 using the MLC algorithm of supervised image classification technique in ERDAS Imagine 2014 software. Fig. 18.9 shows the spatial distribution of LULC in the South Indian coastal stretch. The beach landforms experience the decreasing rate more than 2.0   km2 over time due to the impact of marine and coastal processes and other anthropogenic activities (Kaliraj and Chandrasekar 2012). Therefore, these areas have a high vulnerability weighted value of 5. Sand dunes have seen a significant amount of erosion due to the removal of dune vegetation and human encroachment activities (Huang and Hsieh, 2012), hence these areas are assigned weighted value 4. The cultivable lands show a significant decreasing trend in areas and hence their weighted value is 3. The sandy beaches and dunes increase in areal extent in the up-drift side of the coast and these accreted zones have weighted values 1 and 2. The LULC change is influencing the physical vulnerability to sediment loss or gain along the coastal area for long-term scale.

Figure 18.9. Shows the spatial distribution of LULC in the South Indian coastal stretch.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128196045000184

Land Use/Land Change and Health

Jonathan A. Patz , Sarah H. Olson , in International Encyclopedia of Public Health (Second Edition), 2017

Abstract

Land use and cover change drive a range of infectious disease outbreaks or disease emergences. These drivers include agricultural encroachment, deforestation, road construction, dam building, irrigation, wetland modification, mining, the concentration or expansion of urban environments, coastal zone degradation, and other activities. These changes in turn cause a cascade of factors that exacerbate infectious disease emergence, such as forest fragmentation, disease introduction, pollution, poverty, and human migration. Such changes can subsequently affect biological mechanisms of disease emergence including: altered vector breeding sites or reservoir host distribution; niche invasions or interspecies host transfers; changes in biodiversity (including loss of predator species and changes in host population density); human-induced genetic changes of disease vectors or pathogens (such as mosquito resistance to pesticides or the emergence of antibiotic-resistant bacteria); and environmental contamination of infectious disease agents.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128036785002460

Kenya: A Natural Outlook

Petri K.E. Pellikka , ... Mika Siljander , in Developments in Earth Surface Processes, 2013

5 Discussion: Endangered Ecosystem Services

Land-cover change has numerous ecological, physical and socioeconomic consequences. On the positive side, agricultural expansion may increase food production for a growing population, although it is unsure how productive the last exploited lands will be as they are typically the least favourable. There are numerous negative consequences with both known and unknown links and feedback mechanisms.

Converting the natural vegetation to agricultural land is likely to change the radiation balance of the given unit of area. In principle, the albedo increases as land is without vegetation at least part of the year causing more solar energy to reflect back to the space. Other environmental impacts include the decrease in soil water-holding capacity. As natural vegetation is replaced by agriculture, soil porosity may be reduced by soil compaction, decreasing infiltration capacity and increasing the risks of soil erosion. In mountainous areas, the conversion of the forests to agricultural lands decreases as does the occult precipitation as croplands capture less atmospheric moisture than multilayered indigenous forest or forest of any kind (Holder, 2004). Cloud formation over the land unit also decreases as the evapotranspiration rate is less from fields than from forests causing evidently reduced precipitation.

Further studies project that, by 2030, agricultural land is likely to expand to regions with higher evapotranspiration potential (Figure 6). This expansion will increase by approximately 40% the annual volume of water necessary for irrigation (Maeda et al., 2011).

Figure 6. Histogram showing the cropland patches distribution during 1987, 2003 and 2030, in relation to the historical average potential evapotranspiration in the Taita Hills.

 Adapted from Maeda et al. (2011).

As soil water-holding capacity is reduced, the risk of hydrologic droughts during dry seasons is increased, while during the rainy seasons, soils are more susceptible to erosion. These soil loss and sediment-deposition processes may have a significant impact on agriculture, local economies and ecosystems (Alcantara-Ayala et al., 2006). Although increasing evidences indicate that anthropogenic changes in the landscape are likely to lead to regional and global climate change, the levels and scale of this relationship remain largely unknown. However, it is clear that converting forestland to agricultural land causes changes in local climate via the changes in radiation and water balance. Changes in precipitation and temperature patterns will likely have important impacts on the sustainability of agricultural systems.

The land-cover change taking place in the Taita Hills and its surrounding has been continuing since human settlers arrived in the area and started to convert native vegetation to agricultural land. With ever-growing population and demand for land for cultivation of food crops and other crop types in addition to increase of reserved and protected areas in the study area, land is evidently becoming a valuable natural resource. Land use conflicts have already taken place in the area between farmers, conservationists, settlers and sisal plantation managers (Vanonen, 2008).

Converting natural vegetation, forest or grassland to agricultural areas decreases biodiversity, reduces the capability of vegetation to capture atmospheric moisture and retain water in the vegetation cover, exposes land to be subject to water and wind erosion and changes the radiation balance of the land surface as land is exposed and barren part of the year. These all have still unknown impacts on regional climate.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978044459559100013X

Modeling Bird Responses to Predicted Changes in Land Cover in an Urbanizing Region

Jeffrey A. Hepinstall , ... Marina Alberti , in Models for Planning Wildlife Conservation in Large Landscapes, 2009

Land-Cover Change Modeling

The land-cover change model (LCCM) consists of a set of discrete choice equations of site-based land-cover transitions derived from observed land-cover change (Fig. 23-2) that are applied to geographic information system (GIS) layers to predict land-cover change at a 30 m resolution across four counties in western Washington, representing the central Puget Sound Region. A short description of the model follows; a complete description of the theoretical foundations of the model is available in Hepinstall et al. (2008).

Fig. 23-2. Flow chart of steps performed as part of the land-cover change model and the avian richness and relative abundance models used to predict bird response to changes in land cover.

The LCCM framework derives from the traditions of modeling landscape change as a dynamic interaction between socioeconomic and biophysical processes (Turner et al. 1996, Wear and Bolstad 1998, Wear et al. 1998). The LCCM is written in Python and is designed as a module within the larger Open Platform for Urban Simulation (OPUS) and UrbanSim modeling platforms (Waddell 2002, Waddell et al. 2003, <www.urbansim.org>). UrbanSim consists of a series of modules that have been developed to, among other things, model land-use change in response to changes in transportation networks, household and business location, property development and intensity, infrastructure changes, and policy choices. UrbanSim is designed to aid regional land-use planning.

Urban development models such as UrbanSim predict changes in land use (e.g., undeveloped, residential, commercial, mixed use, timberlands) and development intensity (number of residential units or square feet of commercial space), whereas avian communities respond to changes in vegetation type and structure. We must link models predicting change in land use to models of land-cover change, which then can be used to predict the effects of land development on avian communities. Our LCCM predicts future land cover in response to land-use change and biophysical constraints.

For our implementation of the LCCM, we simulated the potential change to one of eight land-cover classes: heavy urban (>80% impervious surfaces), medium urban (20-80% impervious surfaces), low urban (a mixed class with <20% impervious area and the remaining area in vegetation), grass, agriculture, deciduous and mixed forest, coniferous forest, and clearcut. Each land-cover class can transition from a variable number of other classes. We empirically estimated 26 transition equations as a function of observed land-cover and independent variables from two dates (Fig. 23-2; Turner et al. 1996). The focus of the LCCM is to model urban growth, which in the central Puget Sound is limited to the lower elevations that have little commercial forestland. We chose not to model forest regeneration and instead converted any predicted new clearcut into regenerating forest in the subsequent time step and retained all regenerating forest for the duration of the LCCM run (28 years in this application).

The central Puget Sound implementation of LCCM has multiple possible input dates of land cover to use for developing transition models including 1986, 1991, 1995, 1999, and 2002; we used equations developed from observed 1995–1999 transitions (Hepinstall et al., in press). We modeled land-cover change using discrete choice (multinomial logit) statistical models. Developing a discrete choice equation for each transition modeled is an iterative, semi-automated process that can be done directly within the LCCM code base, but still takes multiple days to complete. Transition probabilities for each 30 m pixel to change from one discrete land-cover class i to another cover class j is potentially influenced by many factors including (1) the predicted type and predicted intensity of a development event; (2) a set of attributes of the pixel; and (3) the land-cover composition and configuration of neighboring pixels (Fig. 23-2; Hepinstall et al., in press). In the Puget Sound implementation of the LCCM, 65 potential explanatory variables are available for specifying discrete choice equations. Land development, or the probability that a pixel will transition from an undeveloped to a developed state, is derived from UrbanSim development module output. UrbanSim output is also used to determine the type (residential, commercial/industrial, mixed use) and intensity (number of residential units or ft2 commercial/industrial added) of development. The remaining variables were developed from spatial databases obtained from county, state, and federal GIS data repositories and required several months to compile and error check. Site attributes influence the ability to develop land through increasing the cost of development (e.g., steep slopes, unstable soils), limiting or prohibiting development (e.g., critical areas such as steep slopes, landslide hazard, riparian areas, etc.; proximity to endangered species habitat), or encouraging development (e.g., proximity to existing infrastructure). Because development events generally occur in patches that are greater than the size of an individual 30 m pixel (900 m2), land-cover transitions in adjacent cells influence the probability of land-cover transitions in a focal cell. The LCCM, therefore, includes distance variables (e.g., distance to central business district) and variables measuring the spatial context of the target pixel, by calculating several measures (e.g., number of residential units added in the previous three years) within 150 m, 450 m, and 750 m moving windows.

The output of the discrete choice equations are probabilities that any given pixel will transition from its current class to one of the possible options for that class including the no-change option. For example, light urban can transition to medium urban, heavy urban, or remain as light urban. Parameter estimates from the discrete choice equations are applied to GIS layers to derive pixel-specific transition probabilities for each pixel to convert to a land-cover class (Fig. 23-2). Because only a small portion of the landscape changes to a new land-cover class over short time intervals (in our case four years), we used Monte Carlo simulations to pick what land-cover type each pixel will be in the next time step (Fig. 23-2). Specifically, transition probabilities for each land-cover class are normalized by the annualized observed transitions and scaled to sum to 1.0 for each possible transition from the starting class. Then predicted transitions are implemented by comparing the class-specific probabilities for each pixel to a random number chosen from a uniform distribution between 0 and 1. If the scaled transition probability to a new land-cover class matches the random value, the transition takes place; otherwise, the grid cell maintains its current land cover.

The LCCM is implemented in the Python language as a component of UrbanSim and can be downloaded and used as a template to develop a local implementation for any region with spatial data for at least two dates of land cover and drivers of land-cover change (i.e., biophysical and socioeconomic). While UrbanSim requires many socioeconomic data layers to fully implement, LCCM is independent of UrbanSim, is flexible, and can be implemented using output from any land-use change prediction.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978012373631400023X

Geospatial environmental modeling of forest declining trend in Eastern Himalayan biodiversity hotspot region

Meelan Chamling , ... Sudipa Sarkar , in Forest Resources Resilience and Conflicts, 2021

6 Conclusion

The LULC changes of the past, present, and future are driven by the random and irrational anthropogenic activities. The predicted LULC categories like built-up area (25.91 km2), agricultural land (4.87 km2), and barren land (2.38 km2) show the increasing trend from 2020 to 2045 (Table 30.7). On the other hand, declining LULC attributes include vegetation (12.97 km2), plantation (12.36 km2), and water body (7.83 km2) between the years 2020 and 2045. The annual rate of change varies from one category to other for mainly unscientific anthropogenic activities. The built-up area will project the highest annual magnitude of positive change of 1.03% from 2020 to 2045, while vegetation attribute faces large-scale depletion with the rate of 0.51% per year during the premeditated period. The area under agricultural (0.19%) and barren land (0.09%) will increase annually. On the other hand, the annual declining trend of transformation will be experienced by plantation (0.49%) and water body (0.31%) between 2020 and 2045.

Although the CA Markov chain model is a competent tool to estimate the future LULC and it can analyze the substantial transformation, yet it possesses several inconveniences; thus, this model is partially ineffective to forecast spatiotemporal LULC change accurately. Since the LULC attributes constantly metamorphose owing to the changing human behavior and unpredicted anthropogenic actions, it becomes a challenging job to predict the future LULC dynamics with machine-learned model. Despite of such difficulty and limitation, dynamic simulating CA Markov chain model has found to be significantly useful for policy makers and administrators to recognize and examine the future LULC change. It proves to be highly useful for decision-cum-policy makers and planners to frame and formulate comprehensive eco-friendly environmental guidelines to conserve and manage the vulnerable but rich biodiversity region at the vicinity of Himalayan mountain range.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128229316000307

Land Use/Land Cover Change Detection and Urban Sprawl Analysis

Cláudia M. Viana , ... Jorge Rocha , in Spatial Modeling in GIS and R for Earth and Environmental Sciences, 2019

29.1.3 Geographic Information Systems and Remote Sensing Techniques in Urban Growth Analysis

Detection of LULC changes is the process of identifying differences in the state of a pixel or phenomenon by analyzing images acquired on different dates (Singh, 1989). Geographic information system (GIS) allows quantification of these changes using remote sensing data, expressed spatially and temporally, to dynamically visualize spatial patterns and LULC composition. These changes can be integrated with social and biophysical data to determine the factors that influence the process of LULC change or its consequences. Often, the resulting linear or nonlinear relationships can be modeled mathematically and statistically analyzed.

Advances in GIS and information technologies have contributed to a substantial increase in research studies since the end of the 20th century, focused on patterns of urban growth and its impacts on human life and natural resources (Terzi & Bolen, 2009). These studies facilitate comprehensive monitoring of physical changes over time. A primary technological breakthrough came from advances in satellite remote sensing (Lo, 1997), from which global projects focused on LULC emerged (e.g., Global Land Cover 2000, CORINE Land Cover, and Land Use and Cover Change). This advanced technology has driven innovative methodologies and better techniques for the classification, monitoring, and time-series analysis of land resources using large archives of satellite data (such as Land [Remote-Sensing] Satellite (Landsat), Moderate Resolution Imaging Spectroradiometer (MODIS), and the SENTINELs). The Landsat collection is an example of freely available and analysis-ready data (for Level 1T, see Hansen & Loveland, 2012; Wulder, Masek, Cohen, Loveland, & Woodcock, 2012), which offers high spatial resolution (30   m) and a large temporal extent (since the 1970s) (Woodcock et al., 2008). New survey and ground-based automated and semiautomated methods have been developed from these advances in remote sensing (Dewan & Yamaguchi, 2009; Maus, Câmara, Cartaxo, et al., 2016). Consequently, spatiotemporal mapping and monitoring can efficiently provide data on multidimensional LULC changes (Lunetta, Knight, Ediriwickrema, Lyon, & Worthy, 2006; Verbesselt, Hyndman, Newnham, & Culvenor, 2010; Weng, 2002; Xiao et al., 2005), enabling quantification of the dimensions and degree of urban sprawl over time (Ewing & Hamidi, 2015; Hamidi & Ewing, 2014; Jiang, Ma, Qu, Zhang, & Zhou, 2016; Sarzynski, Galster, & Stack, 2014; Weng, 2001; Zhang et al., 2016).

Classification of satellite images is considered a complex and time-consuming process. Moreover, classification accuracy can be affected by many factors, such as the type of input images, the selected classification methods, and the algorithm applied (Lu & Weng, 2007). Algorithms and other such tools that prioritize the time dimension and are capable of spatiotemporal analysis of remote sensing imagery have been used effectively to identify urbanization process dynamics (Seto & Fragkias, 2005).

Despite these efforts, some challenges regarding LULC classification based on time-series persist (Petitjean, Inglada, & Gancarski, 2012), namely: (1) the irregular temporal phenological signatures of different land cover classes (Maus, Câmara, Cartaxo, et al., 2016; Petitjean et al., 2012); (2) the insufficient sampling used to train the supervised algorithms; and (3) the missing temporal data. The dynamic time warping (DTW) method (Guan, Huang, Liu, Meng, & Liu, 2016; Petitjean & Weber, 2014; Petitjean et al., 2012; Rabiner & Juang, 1993) proved capable of dealing with these challenges (Baumann, Ozdogan, Richardson, & Radeloff, 2017; Guan et al., 2016; Maus, Câmara, Cartaxo, et al., 2016; Petitjean & Weber, 2014; Petitjean et al., 2012). More recently, Maus, Câmara, Cartaxo, et al. (2016) developed the time-weighted DTW (TWDTW) algorithm, improving on the DTW. TWDTW balances shape matching and temporal alignment, allowing both image classification and spatiotemporal analysis. It represents one of the few existing open-source methods for remote sensing time-series analysis, and is freely available through the dtwSat package (Maus, Câmara, Appel, et al., 2016) and related graphical and statistical R software tools (dtw, proxy, zoo, caret, mgcv, sp, raster, and ggplot 2) (Maus, Câmara, Appel, & Pebesma, 2016). Maus, Câmara, Cartaxo, et al. (2016) have demonstrated that this method achieved high accuracy for LULC classification from a MODIS enhanced vegetation index (EVI) time-series. Furthermore, Belgiu and Csillik (2018) show that TWDTW achieved a high overall accuracy for LULC classification using a Sentinel-2 NDVI (normalized difference vegetation index) time-series.

Since urban growth is a spatially conditioned process, where an event at a given site is partially affected by occurrences in neighboring locations, the TWDTW method was used to classify urban areas over a long time-series computed from Landsat satellite imagery. The classification of the images was based on computation of four spectral indices, namely: (1) NDVI; (2) normalized difference built-up index (NDBI); (3) normalized difference bareness index (NDBaI); and (4) normalized difference water index (NDWI). Other studies have also used indices to identify the built-up and bare land in urban areas, for example, NDBI (Zha, Gao, & Ni, 2003), index-based built-up index (Xu, 2008), urban index (Kawamura, Jayamanna, & Tsujiko, 1996), (NDBaI) (Zhao & Chen, 2005), and bare soil index (Rikimaru & Miyatake, 1997). To improve the differentiation between urban and other LULC classes, the satellite data were mapped in a three-dimensional array in space-time (Maus, Câmara, Appel, et al., 2016), from which the four spectral indices were derived.

This study presents the results of applying the TWDTW method, using R tools, for LULC detection and urban sprawl analysis. The patterns and processes of urban growth for a test area in southern Portugal are identified and discussed.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128152263000296

Context and background of urban heat island

Ansar Khan , ... Yupeng Weng , in Urban Heat Island Modeling for Tropical Climates, 2021

1.8 Book design and methodological overview

Urbanization and LULC changes have led to not only brisk growth of the global economy but also a substantial increase in urban population due to migration, which has brought in many acute problems in the urban environmental system. Among these growing concerns, UHI effect has been widely considered as a significant urban environmental dilemma, since it causes a wide range of detriments to urban life. The present UHI study has been conducted in the case of the KMA region for the purpose of proper scaling of urban thermal fields and harnessing UHI countermeasures particularly applicable for tropical cities. In this regard, the study has adopted systematic review of existing literature to find gaps in tropical UHI researches carried out so far by adopting descriptive approach (survey and documentation), correlation techniques (based on temperature observations) along with semiexperimental (field experiment), and experimental (WRF and ENVI-met) methods. For mapping and modeling purposes, the field meteorological data, optical and TRS data (Landsat 7 and 8), and global forecast system (GFS) data and other relevant and ancillary data were used. Finally, a data assimilation approach has also been employed on processed data to prepare subsequent data as required. The following paragraphs present the research design and overview of methodology that have been adopted for achieving research goals.

•

For characterization of thermal fields and evaluating UHI effects, the study uses monowindow algorithm (MWA) for retrieval of LST and derivation of urban heat island effect ratio (UHIER) to assess seasonal characters of urban thermal fields. Furthermore, matrix of transition probability was computed to examine persistence or dynamics thermal field types during the period 1996–2017. The UHI thermal field elements have been derived by using fractal net evolution approach (FNEA) applied on LST rasters. The spatial concentration of UHI thermal fields was extracted by Getis-Ord index ( G i ∗ ) that could delineate the cold and heat island areas. The evolution of UHIs in Kolkata region is quantitatively assessed by computing metrics for each temperature class and also for the entire thermal landscape using FRAGSTATS application.

•

For realization of the urban environment and UHI drivers, this research first develops LULC classifications of the study area at decadal intervals by applying support vector machine (SVM) algorithm on training data sets. Then an effort has also been made to show a relationship between rate of LST and LULC change for each parcel of the urban landscape of KMA. Furthermore, on the basis of derived results, major UHI drivers have been identified from three broad categories—socioeconomic (population distribution and LULC), structural (building materials, building height, building age, building association), and radiative drivers (building roof type, building roof reflectance, and road pavement). Finally, the contribution of built-up areas to modify urban thermal character has been quantified pursuing a more generalized approach through a new and simple method, i.e., stepwise land-class elimination approach (SLEA).

•

For scaling of UHI zones and microclimate environment, thermal parameters (summer temperature, interseasonal temperature difference, and seasonal trends of temperature) have been derived from LST rasters and used for a fuzzy-based classification approach in e-Cognition platform. The condition values for fuzzy membership functions have been fixed by normalizing the LST scenes according to suitable probability functions. The limiting values thus computed were used to classify each raster of the thermal parameters into segments with high, medium, and low values for application in the fuzzy process. Thus, the KMA has been classified into six major UHI zones and microclimate environments on the basis of fuzzy classification—(1) steady and intense UHI, (2) emerging and intense UHI, (3) potential UHI with moderate intensity, (4) moderate potential UHI with lower intensity, (5) less potential UHI with lower intensity, (6) balanced thermal field, and (7) unclassified. Those areas of KMA that did not fall under applied conditions have been brought under unclassified UHI areas. The energetic balance differentials, mean center, and directional growth of each zone are also analyzed. Finally, derived UHI and microclimate zones have been validated with the universally acknowledged and accepted scheme of local climate zones (LCZs).

•

For simulating UHI effects at the city scale, WRF/UCM coupled model was adopted. The study first assesses the model performance and validates the model-derived results with observed temperature. Then, the impacts of urbanization and LULC conversion on urban temperature change have been analyzed on the basis of up-to-date LULC data and observed values of urban physics parameters (energy balance, wall and roof temperature, radiation, and momentum flux) were derived for the model domain. Finally, UHI intensity and magnitude were simulated for future scenarios.

•

For simulating the thermal effects of UHI mitigation strategies (cool roof, cool pavement, urban greening, and a combination of the three) at the microscale, the ENVI-met model has been used for three building environments selected from the study area. The simulations were performed for 30-h using in situ data (weather, vegetation, soil, and building) to experiment with the strategies over a complete diurnal cycle. The results could be useful for planning energy smart and sustainable cities in tropical conditions ensuring thermal comfort for the urban dwellers.

Cities could be the best laboratories for researching the impacts of global climate change since much of the climate change risk is concentrated in urban areas, particularly the cities in tropical regions that are likely to see the strongest impacts from climate change. However, research on the augmentation of climate change effects by local urban warming (herein UHI) remains weak for tropics. Much of the climate agency overlooked the role of tropical urban areas both as a forcing factor and as a key stakeholder in managing climate risk and crisis.

In general, a key difficulty in untangling the tropical urban warming from global warming–induced climate change is the computational and parametric challenges associated with representing urban areas in different high-resolution climate models without modeling the urban areas themselves, a technique not without problems. The insufficiency of data set (necessary for model parameterization) that is captured for the fast-expanding tropical cities (as most of cities in tropics are located in the developing countries) is a major hindrance in conducting the model-based research. Moreover, the rapid and haphazard process of urban expansion in the developing countries and the fast-changing surface characters by means of concretization adds further uncertainty and challenges in such modeling. Hence, the dynamics of LULC in the cities of developing countries needs to be adhered while modeling the UHI effects.

Although the situation is continuing to improve, still more efforts need to be made to (1) reduce UHI effects and (2) implement the popular mitigation strategies as a part of climate risk adaptation. Thus, the mitigation of UHI effects in the tropical urban areas is surely weak on account of two technical facts: (1) knowledge of UHI in tropics still remains patchy and numerically weak, and (2) the proliferation of strategies focusing on heating-only climates does not readily translate to cooling-only regions.

The distinguishing nature of the tropical UHI is attributed to two specific features of surface energy balance (SEB) in the urban areas (Giridharan & Emmanuel, 2018). Firstly, the predominance of direct solar radiation in net all-wave radiation due to the high angle of the sun over the tropics induces excess heating of the urban structures (Box 1.3) in contrast to the midlatitudes (Chow & Roth, 2006). Secondly, during the dry season, the tropical areas receive greater amounts of net radiation, and the latent heat transfer remains considerably higher than the sensible heat transfer, while both the conditions reverse when the rainy season sets in (Barradas, Tejeda-Martı́nez, & Jáuregui, 1999). The unusual premonsoon daytime negative UHI in tropical cities is explained by low vegetal cover in nonurban surroundings that limits the conversion of net insolation into latent heat (Emmanuel et al., 2007; Oke, 1980; Shastri et al., 2017). The shading effects of plants and buildings are also responsible for lowering the temperature in urban locations of the tropics on dry summer days (Emmanuel, 2005; Johansson & Emmanuel, 2006; Johnson, Kovats, McGregor, Stedman, Gibbs et al., 2005). Therefore, selection of the most appropriate measure to effectively mitigate tropical UHI will depend on a proper understanding of the SEB mechanisms for this region. Therefore, modeling the relationship between the building environment and its surrounding outdoor areas is of critical importance for guiding the urban climate and outdoor thermal comfort (Rizwan et al., 2008).

Box 1.3

Urban heat island in tropics

The heat retentive engineering structures, arrangement of a city's streets and buildings play crucial roles in the local urban heat island (UHI) effect in any big city irrespective of climate type, which cause cities to be hotter than their surroundings. But the tropical cities are more susceptible to climate effects of urbanization for the following reasons.

•

The effects of global warming are expected to be strongest in the tropics due to more direct sunshine over this region. Therefore, the warming of cities through urban growth may aggravate global climate change effects to an unbearable extent.

•

Heat waves are very common weather phenomena found to occur in many tropical regions during summer. UHIs are the potential for aggravating the heating effects of heat waves.

•

Tropical trees with greater leaf surface area and extensive root zone transfer significant amount of water vapour from surface to the atmosphere in terms of evapotranspiration and thus facilitates surface to atmosphere heat transfer in from of latent heat. The latent heat transfer constitutes a significant component of surface energy balance (SEB) in tropical cities. Therefore, the replacement of green cover by impervious surface upsets SEB in tropical climate.

UHI has become a growing concern to the quality of densely built urban environments, particularly in tropical cities. In tropical climate, urban geometry modifications could be helpful through installation of radiant cooling strategies, building ventilations, and evaporative cooling surfaces.

The climate-sensitive design of buildings that enable climate change adaptation is required to combat UHIEs. In tropics the cooling load is the principal design problem, and it can be overcome through different high-resolution modeling approaches at neighborhood scale (building scale) rather than at the city scale (urban scale).

The impacts of climate change in urban areas have become a significant and serious aspect of UHI climatology. However, for countermeasures of UHI phenomena, a series of climate-sensitive policies and scientific strategies were proposed and implemented in the urban areas of temperate climate, whereas tropical cities are still far from such types of countermeasures. Moreover, existing rural–urban ( Δ T = u t − r t ) difference (between urban and cropland or rural environments) does not imply the actual UHI magnitude. Hence, a proper scaling of UHI zones should be considered for measuring actual amplitude of UHI. In the present study, UHI magnitude has been computed on the basis of normalized values of temperature across LST scenes and rasters for temperature parameters derived from LST. This technique reflects the actual magnitude of UHI in KMA areas. The study also employed the modeling approaches (deterministic and stochastic) at the building scale and city scale to evaluate the effectiveness of popularly used strategies for urban cooling in addition to predicting the UHI effects. The derived results reveal that the compatibility of each type of UHI countermeasure differs from the other with respect to building environment and scale of application. Although, for modeling of large-scale UHI effects, the accuracy is not enough to predict precisely, spatially and computationally more efficient models are to be required for future UHI research. Hence, the scale and climatic environment are the crucial aspects in UHI studies.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128196694000015

Land Use and Land Cover Transition in Brazil and Their Effects on Greenhouse Gas Emissions

Eráclito R. de Sousa-Neto , ... Jean P. Ometto , in Soil Management and Climate Change, 2018

Land Use and Land Cover Change Science

Land use and land cover change science were proposed to better understand the anthropic impacts on the earth systems through changes in land management and to observe such changes by satellite imagery, by studies of drivers and impacts and by modeling the processes of land change, looking to provide a better picture of the land transition dynamic and scenarios for future changes ( Robinson et al., 2013). Furthermore, land change science is linked to studies and estimates of GHG emissions, as changes in land cover lead to alterations in soil properties and consequently to changes in nutrient dynamics, which are responsible for gaseous fluxes.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128121283000203

Human Effects on Climate Through Land-Use-Induced Land-Cover Change

Andy Pitman , Nathalie de Noblet-Ducoudré , in The Future of the World's Climate (Second Edition), 2012

4.2 The Scale of Human Modification

Humans have undertaken intensive LULCC at a scale commonly underestimated by scientists working in other fields of climate science. Several groups have undertaken reconstructions of the scale of LULCC – work that underpins current efforts to determine the impact of this change on climate (e.g., Defries et al., 1995; Ramankutty and Foley, 1999; Klein Goldewijk, 2001; Hurtt et al., 2006; Pongratz et al., 2008). Figure 4.1a shows that by 1500 large areas of Western Europe had been partially cleared for agriculture and for timber. LULCC intensified, particularly in Western Europe, through to 1800. Indeed, by 1750, 7.9–9.2 million km2 (6%–7% of the global land surface; note that all these percentages are calculated over land excluding Greenland and Antarctica) were in cultivation (Forster et al., 2007) although only in Western Europe and perhaps parts of northern China had the intensity of LULCC led to more than ~60% agricultural cover for a given region (Figures 4.1b, 4.1c, 4.1d). By 1990, 45.7–51.3 million km2 (35%–39% of global land surface) was in cultivation, forest cover had decreased by about 11 million km2, and intensive LULCC had impacted parts of the USA, much of Western Europe, India, northern China, and elsewhere. Large areas of the Southern Hemisphere underwent LULCC throughout the nineteenth century. By 2000 (Figure 4.1f) the fingerprint of human activity through LULCC has only not affected a few desert regions, the central Amazon and Congo Basins, and the Arctic and Antarctic (not shown). Figure 4.1 needs to be interpreted qualitatively. A careful examination points to some changes, such as over Australia between 1800 and 1900, which seem unlikely, but the overall pattern of changes are probably reliable at large scales. Williams (2003) provides a detailed account of these global and regional changes as well as their underlying causation.

FIGURE 4.1. Reconstructed and projected LULCC for various time periods. The scale is the relative fraction of any grid box containing the sum of pasture or crops.

 (Source: These data were downloaded from the Land Use Harmonization web site at http://luh.unh.edu ).

The easiest and so far most traditional way to quantify the impact of these changes is by means of the surface albedo (the fraction of incoming solar radiation reflected by a surface), which can be translated into a change in radiative forcing. Forests are dark to the wavelengths of visible light and absorb very efficiently (albedos range from around 0.05–0.2). Croplands are commonly much more reflective (around 0.2–0.25 when they are snow-free, but much higher in the presence of snow: Bonan (2002)). Thus, the global net seasonal and annual impact of LULCC is an increase in albedo. While there is considerable uncertainty in the precise impact of LULCC on radiative forcing, Forster et al. (2007) suggest a reduction of 0.2 W m−2 ± 0.2 W m−2 on the global average. While this appears to be a large range (perhaps as large as 0.4 W m−2 or as small as 0.0 W m−2), available evidence points to it being very small compared to well-mixed greenhouse gases (GHGs) (about +2.5 W m−2).

Reasoning of this kind has led to the role of LULCC being ignored in most climate projections. The reports by the Intergovernmental Panel on Climate Change (IPCC, e.g., the Fourth Assessment Report, Solomon et al. (2007)) highlight LULCC as an area of uncertainty but the model simulations assessed in the Fourth Assessment Report do not include LULCC in terms of how it modifies the surface (only direct emissions of CO2 are accounted for). Pitman et al. (2009) discuss the implications of omitting LULCC from the IPCC assessments and conclude that at the global-scale there is no evidence that this matters to global metrics, although a more recent study by Davin and de Noblet-Ducoudré (2010) demonstrated that oceanic feedbacks have the potential to enhance the LULCC impacts. Overall, IPCC AR4 statements of the likely amount of global warming associated with a given emission scenario are not flawed globally due to the lack of inclusion of LULCC. However, while GHGs are globally well-mixed, LULCC is highly concentrated at present in Western Europe, the eastern US, China, and India (Figure 4.1f). Thus, while the global impact may be negligible, and the impact on global means may be small, the global impact is contributed from a fraction of the land surface strongly coincident with human population (cf. Cleugh and Grimmond, 2012, this volume). Furthermore, it has been pointed out (Pielke et al., 2002; Davin et al., 2007) that radiative forcing is not a good way to measure the impact of LULCC on surface climate. A regionally-significant impact on climate can be achieved by LULCC without any change in albedo if the partitioning of energy or rainfall is modified. LULCC changes the seasonality of heat and moisture fluxes, changes the probability of extremes, and may provide a local perturbation that triggers a change in the atmosphere sufficient to cause changes remote from the perturbation. Thus, while the global impact of LULCC on mean radiative forcing and mean global climate sensitivity may well be negligible, this finding is really only relevant in theoretical studies since humans and the Earth's terrestrial flora and fauna live, source their food, and source their water at local and regional-scales.

The question is therefore not whether LULCC affects the global climate (it does via release of, for example, CO2; see Dickinson, 2012, this volume). Rather, it is whether it affects climate at any space or timescale of significance for living bodies; and clearly 'significant' is a value judgement. A major change in climate over one region may be insignificant to the global climate scientist, but highly significant to the affected population.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978012386917300004X